Dynamic Pricing Strategy Runs on Three Signals, Not Just the Market with Mridul Bansal

Summary and Highlights

🎙️ Who Is Mridul Bansal

Mridul Bansal is a Product Marketing Manager at PriceLabs, the revenue management platform used across hundreds of thousands of listings worldwide. He has been with the company for close to three years.

His role sits in an unusual spot. As he puts it, he is the voice of the customer inside the company, and the voice of the company back out to customers. That means constant conversations with hosts and property managers, feeding what he hears into the product roadmap, then making sure operators actually know what got built and how to use it.

Mridul is based in India and speaks regularly on revenue management, pricing strategy, and data driven hosting at industry events and webinars. He is a firm believer that market data should inform decisions rather than replace judgment.

The part of the job he loves most is the people. A pilot who runs a rental. A lawyer. A school teacher. All connected by the same business.

---

📊 The Dynamic Pricing Strategy Behind Every Rate Recommendation

This was the heart of the episode. A solid dynamic pricing strategy is not one calculation. Mridul broke it into three layers that stack on top of each other.

Market data. PriceLabs pulls publicly available listing data from Airbnb, Vrbo, and Booking.com. ADR, booked rates, active listing counts, and pacing against the same week last year. The algorithm reads demand signals without always knowing the cause. When rates in a market spike 60 percent for one weekend, the system reacts even if nobody has told it a concert is coming to town.

Your own performance. Occupancy and pacing compared against your history and against the market. Mridul used a sharp example here. Say December occupancy sits at 80 percent versus 90 percent last year. That looks like a decline. But if the market is down 20 percent, you are outperforming. Your rates may not need to move at all.

Your preferences. How aggressive you want to be. Whether you would rather fill every night on thin margins, or hold rate and accept empty nights. There is no universally correct answer here, and Mridul was clear that the tool is meant to reflect your goals rather than override them.

Those three combine into the recommended price. If you are still setting rates on gut feel, the systems first approach to scaling a short term rental business is a good companion read.

---

🔬 Anyone Can Get the Data, Almost Nobody Can Clean It

Gil pushed on something interesting. Operators in Facebook groups are vibe coding their own pricing tools, scraping public data and building a model.

Mridul’s response was more generous than expected. He confirmed the data really is public, and that PriceLabs does not use one customer’s private data to price another. In theory, you could gather the same inputs.

The moat is what happens next.

Owner blocks that look identical to bookings. Duplicate listings. All inclusive rate displays that hide the real nightly figure. Commission baked into a booked price, so a $300 booking yesterday cannot become a $300 recommendation today. Twelve years of nuance sits inside that cleanup.

“It’s not just about gathering the data. It’s really about treating that data right.” That line stuck.

---

🗺️ Why a Radius Around Your Property Is the Wrong Map

The hyperlocal piece was the most practical stretch of the conversation.

Draw a circle around your property and you will pull in comps that are not comps. Mridul used a lake example. Properties on the far shore show up inside your radius, but no guest is weighing your home against theirs.

Gil matched it with a real client market. Lake Anna has a warm side and a cold side. The warm side is private and not open to the public. Pricing between the two is dramatically different, and there is no amenity checkbox in your PMS that captures it.

The same logic applies to a stadium. A property near the venue and one ten miles out are living in different markets on concert weekend.

If your comp set is wrong, every downstream decision is calibrated off. This is the same specificity that makes SEO landing pages for vacation rentals work, and the same reasoning behind investing in amenities that actually move revenue rather than generic upgrades.

---

📱 The Mobile App Was Built to Solve Two Problems

Mridul’s proudest recent launch is the PriceLabs mobile app, and he was deliberate about what it does not try to be.

Problem one is quick edits. Change pricing from wherever you are, in seconds, without opening a laptop. He mentioned one customer who adjusted rates from a yacht, which he found both flattering and slightly concerning.

Problem two is awareness. Plenty of hosts have automation running correctly and simply want to glance at what this weekend is priced at, or Christmas, or the week before Thanksgiving.

Gil’s own use case landed here. Thanksgiving moves every year, and Smokies pricing for those dates needs attention well in advance. Then there is the last minute window inside 14 days, where discounting and length of stay changes need to happen at the pricing layer rather than inside Airbnb, or every other channel falls out of sync.

Cramming the full desktop experience into a phone would have made it worse. They resisted.

---

🤖 What the MCP Connector Actually Changes

PriceLabs launched an MCP connector alongside the app, which means you can work with your revenue data inside Claude instead of logging into a dashboard. Gil covered similar ground when Hospitable shipped their own MCP server, and the pattern is spreading fast across STR software.

Mridul’s favorite prompt is the one hosts have always wanted to ask. Why are my prices down this weekend?

The answer comes back with your performance, market performance, and the thresholds you set, then follows with what you could do about it. No graph reading required.

Gil described a friend who runs an automated morning report. Year over year comparison, anomalies, and the last seven days of bookings. It functions as a pulse check that tells him whether the business needs his attention today or not.

That habit matters more than the tool. Most operators lose weeks to the wrong priority simply because nothing told them where to look.

---

🪜 The Third Tier Most Operators Have Not Entered

Gil framed a maturity ladder that is worth sitting with.

Tier one is flat or manual pricing set by instinct. Tier two is a dynamic pricing tool running inside its dashboard. Tier three is a dynamic pricing tool augmented by AI, where the operator queries their data conversationally and acts on it proactively.

Dynamic pricing itself is no longer a differentiator. It has become table stakes. The gap now sits in how everything else in your operation connects to it.

Mridul added a useful caveat. Tier three is a choice, not a requirement. A host with three properties who has configured their preferences well and lets automation run is not doing it wrong. For a revenue manager overseeing hundreds of units, where this is the whole job, the calculus changes completely.

Dan Rivers made a similar point about treating revenue management as a full system in his episode on the Airbnb algorithm, and Tim Hubbard covered the strategy layer while scaling 220 properties remotely.

---

🛠️ What You Should Not Build Yourself

Gil was direct on this. Learn to use Claude and the AI tools available to you. Also know where the line is.

Dynamic pricing, your property management system, and your booking website all sit on the far side of that line. He reviews vibe coded sites that hosts proudly share and finds the fundamentals missing under the hood. Na’ím Anís Paymán raised the same quiet risk in his episode.

A site can look finished and still be invisible to search. Server side rendering, schema, and structured content are what get properties surfaced by ChatGPT and Google, as covered in optimizing STR SEO for Google, ChatGPT, and AI search and the GEO framework for vacation rentals.

Gil also shared where CraftedStays is heading with Aria, which scans a portfolio and the web, then tells operators which content they are missing. Someone with no SEO background learns they should have a landing page for large group stays or pet friendly properties. That research shows up in why local operators are beating national brands in AI search.

Mridul’s reaction was telling. Marketing affects every host whether they want to learn it or not.

---

🪧 Use OTAs as Billboards, Then Build the Bridge

Asked for one tactical direct booking move, Mridul gave the clearest version of the billboard effect we have heard on the show.

Stay attractive on OTAs. Competitive pricing, strong photos, well written descriptions. That is your billboard doing its job.

Then build the bridge. Put your brand name in the listing description, and in the title if you can. His test is brilliant in its simplicity. Show your OTA listing to a stranger and ask them to find your direct booking site. How long does it take?

Give people a reason to cross the bridge. A better rate, or a small add on. A bottle of wine for a couple celebrating something. Flowers. Gestures that cost little and signal a lot.

Patryk Swietek runs the same brand naming play in Joshua Tree, and Amber Knight explains why traffic alone will not fill your calendar once guests land.

Gil closed with a story. A guest wanted his property but had a fixed budget. Told the fees and taxes could not be discounted, they read between the lines, emailed, and booked direct under budget. He also finds direct guests are simply better guests. Once you have that relationship, keeping it alive through email is what compounds.

One note from Gil worth repeating. Keep the brand mention in your listing and photos. Do not put it in your OTA messages.

---

⚡ Rapid Fire With Mridul

📚 A book that inspired you?
Shoe Dog by Phil Knight. Mridul read the Nike memoir young and it stayed with him. Not because it relates to his work, but because of how honestly it shows what building a company costs, including the losses along the way.

🧠 One piece of mindset advice for someone starting something new?
Do it for the process. Enjoy what you are doing every day rather than waiting on a golden outcome. Have the goal, but let the daily work be worth something on its own.

🎯 One tactical direct booking tip they can act on this afternoon?
Stop framing it as OTAs versus direct. Treat OTAs as billboards, keep your listings genuinely attractive, put your brand name where a guest can find it, and give them a reason to book on your site once they do.

---

🔗 Connect With Mridul Bansal

💼 LinkedIn: linkedin.com/in/bansal-mridul
🌐 PriceLabs: hello.pricelabs.co (30 day free trial, no card required)

Mridul shares revenue management and pricing insights regularly on LinkedIn, including the MCP prompts he mentioned in this episode.

---

🎧 Listen, Then Come Build With Us

This conversation gives you the mental model behind your rates. The full episode has more, including how PriceLabs simplified onboarding into a five step setup and where their AI assistant work is heading.

🎙️ Listen to the full episode of the Booked Solid Show on Spotify, Apple Podcasts, or YouTube.

And if you want to be in the room where these conversations keep going, join the CraftedStays community of operators building direct booking brands. Real hosts, real numbers, no gatekeeping.

Your pricing tool tells you what to charge. Your website decides whether anyone can book it from you. If your traffic is not converting, or you have been quoted thousands for an agency build, there is a better path.

👉 Start your free trial at CraftedStays.co and build a direct booking site that turns your billboard traffic into your bookings.

Transcription

Mridul: I think one of the important parts of dynamic pricing is that it doesn’t work just based on market data. If you were to just blindly follow market data, I don’t think that’s the way it works. It also understands your occupancy, your performance in general.

So maybe for the month of December you are booked, your occupancy is 80 percent, whereas last year in December your occupancy was 90 percent. Clearly you would understand this as a down period where you are making less than you were, or you’re less booked than you were last December.

But when you look at market data, the market is actually down by 20 percent. So you being down 10 percent is actually better. You’re still outperforming the market. That’s where your factor comes in as well. Let’s say for next weekend the market is quite down, but you’re already doing good enough. You’re already good as per your past data, your market. So your prices don’t necessarily have to go down. Even if the market is down, if you are doing good enough, you can still stay at the same price and just hope to get booked. Otherwise, you’re still doing good enough.

Gil: Before we bring on our guest, I want to talk about something I’ve been hearing a lot from hosts. I keep hearing the same thing: “I know my website isn’t converting, but I can’t afford $8,000 on an agency to rebuild it.” Here’s the thing, you’re learning all these marketing strategies, you’re driving traffic, and you’re putting it all to work, but if your site isn’t built to convert, you’re basically lighting your energy and money on fire.

Even if you could afford an agency build, every time you want to test something or make a change, you’re having to pay them again. You can’t iterate, you can’t test, and you really can’t improve on things. You don’t need a custom $10,000 website to get the conversion rates that matter. You just need the right platform.

That’s why I built CraftedStays. It’s purpose built for short term rentals and designed from the ground up to help you drive more direct bookings. You can finally turn that traffic into bookings, and you can keep testing and improving as you learn. You can make changes all on the platform. You don’t need to learn something new.

So if you need help or you want to get started, go to craftedstays.co and start your free trial. Now let’s bring on our guest and get deep into hospitality and marketing.

Hey folks, welcome back to the Booked Solid Show, the show where we bring top operators to discuss hospitality, operations, and direct bookings.

On today’s show I have Mridul Bansal from PriceLabs. He’s a product marketer and works really intimately with the product team. Today he shares the more intimate parts of PriceLabs, including why they decided to create a mobile app, the usage of MCP to level up your revenue management game, including what MCP is, and his role in making sure that property managers are making the most out of the important things PriceLabs is building.

So without further ado, let’s bring him in. Hey Mridul, welcome to the show.

Mridul: Hey Gil, thanks for having me. Excited to be here.

Gil: I’m excited to have you here. I’ve met a lot of folks from the PriceLabs team over the last few years, especially at all the conferences. I’ve met Sreeja, I met Richie. But I especially wanted to have you on from the product marketing team because you are so intimately involved in it. Having someone that is very tactical working with the product team, figuring out what challenges people have and how they’re thinking about dynamic pricing, that’s a wealth of value. So I appreciate you being on the show.

Mridul: Great to hear, Gil. Thanks.

Gil: I gave a short introduction on who you are and what you do, but for folks that don’t know about you, can you introduce yourself?

Mridul: Of course. I’m Mridul. I am based out of India. I currently work as Product Marketing Manager at PriceLabs. I’ve been with the company for almost three years now, and my job is basically to be a voice of customer for the company and a voice of company for the customer. I speak to a lot of hosts and property managers and get insights, get their thoughts on how we can do better.

What are the products we can build? How can we improve our existing products? And at the same time, whenever we launch a new product, whenever we add a new feature, I go to our customers and tell them about it so they can get the most value out of PriceLabs. That’s pretty much in a nutshell what I do.

I think the best part about my job is talking to hosts and property managers. I love this industry because of the people. Last week I was talking to someone who’s also a pilot, and he runs a short term rental. In one of my earlier masterminds, one was a lawyer, one was a school teacher. You meet people from so many diverse backgrounds, and they’re all connected because of hosting and short term rentals. I love that.

Gil: I think we have a very intimate space where folks come from very different walks of life. They come into short term rentals maybe as an investor and they end up growing their portfolio. I get thrilled every time I see someone say, “I’m leaving my W2 because my short term rental income has given me more than enough to survive and thrive.” So I totally agree that it’s a very dynamic space we’re in.

Mridul: Exactly.

Gil: One of the challenges in the software space, and I’d love your product marketing feedback on this, is that oftentimes we are shipping out new features and our customers are using them, but your role is making sure that the time and energy the team is spending actually gets to the customer’s hands and they’re using it.

For folks that don’t know what product marketing is, in the context of PriceLabs, what does your day to day look like in terms of making sure that the products PriceLabs produces end up getting awareness with consumers?

Mridul: One reason my job gets a bit easier is that it all starts with customer feedback. We have a product, and we talk to customers, we talk to people who are using it day and night. We talk to people who’ve been using it for a while and people who just started using it.

We really try to observe and understand what improvement they would like to see. It’s not just about what our team thinks should be added, but based on actual customer feedback on what makes sense for a customer at this stage to have. It all starts with that.

In my day to day, I generally sit with the product team and look at all the numbers and data. This is how we see our customers using a particular feature. This is where we see most clicks happening. This is where we see some drop offs happening, which means maybe customers are not able to go through this journey in the product.

For example, at PriceLabs, let’s say we have a button called Review Prices. We would monitor how many people are clicking on that button and then not going through. If the number is not healthy enough, that means there’s something that needs to be fixed. So we plan these product roadmaps, and then it actually goes into development. A lot of cycles of designing and development. A lot of magic that the engineers do, which I never understand. And eventually we get a feature or a product.

What we do from there as part of product marketing is to launch it. We generally have a beta version where we invite some of our customers to try that feature or product. Before giving it to the entire customer base, we get feedback from beta users on how they are perceiving it. What are we doing right? What can we improve? Based on their actual feedback, we keep iterating until it is ready to go all in.

Once it is ready, we let our customers know. There are different channels for it. We email our customers, maybe we show a pop up inside the PriceLabs dashboard. So if someone is a bit irritated because of those pop ups, sorry, that’s on me. Then we post about it on our LinkedIn, on our socials, and just try to let everyone know that we have this feature and they can come try it.

Gil: That’s an important thing to know about this specific role. Your job is to make sure that the users of PriceLabs are maximizing the ROI they’re getting from the platform, that they’re using the features that get developed.

Far too often on the product engineering side, we produce features because we think it’s going to help our customers. Maybe several customers have been asking for it, so the engineering team prioritizes it and puts it on the roadmap. But eventually it gets to a point where it’s in the wild, and if folks don’t know about those features, it ends up becoming a big loss because it’s a pretty heavy engineering investment.

This is an extremely important role. I was a product manager for 15 years prior to starting this company, and I worked very intimately with the product marketing team. That was one of the strongest relationships I had in the org, because without them we would be developing features that never get used.

Mridul: You’re right, Gil. I completely agree. Partly because it makes my job look cool, but also because I do agree it’s a very important part, to actually let your customers use whatever you have built. Otherwise it all goes in vain.

Gil: And I bet every feature doesn’t deserve the same sort of announcement or marketing on your side. As you think back at some of the wins over the last three years, what is one product launch you worked on that was received phenomenally well, that got really good adoption, and that you were really proud of at PriceLabs?

Mridul: That’s a great question. I think I do have recency bias, because very recently we launched our mobile app. Now you can access PriceLabs on your mobile. We are intentionally keeping it light. At least at this point, we don’t want to give all the features of desktop on mobile, because we want it to be something you can just use wherever you are.

That has received great traction and great feedback. We have all five star ratings everywhere. We literally have people reaching out to our team, our solution consultants, our support team. Everyone is reaching out to us on different channels talking about how they love the app.

Even in terms of numbers and interactions, we have a large number of users who have already downloaded the app and are using it daily. That’s something I’m very proud of. The mobile app was a product I was really excited about, and I was very happy that we were able to convey that to our users and that they find value out of it.

Gil: Talk to me a little bit about the mobile app. I’m guessing it’s not everything you can do on the web dashboard. It’s a very tailored experience on the mobile device. What were the biggest problems you were trying to solve when the PriceLabs product engineering team decided to tackle this?

Mridul: We wanted to solve two things essentially with the mobile app, and I’m sure it will evolve as the app keeps going into the hands of our customers. But at this point, we wanted to solve two things. One was: how can I, as a customer, as a host, as a property manager, make quick edits?

Sometimes you’re not in your office, sometimes you are somewhere out. I remember I spoke to a customer who said that last weekend he made a price change when he was literally on a yacht, and I was like, “That’s great. I don’t think you should be using your phone while being on a yacht, but I’m happy to know that you used it.”

So one of the first reasons we built this app was to give you that freedom to make edits to your pricing anywhere you are. You don’t have to log into your laptop. You don’t have to sit at a desk. You can just, on the go, open an app and make an edit, which literally takes seconds.

The second reason we wanted to build the app was to keep you aware of your pricing. We have heard from a lot of hosts that, “Okay, I know my pricing is working. I know it is automation, I have set up my rules and it all works, but maybe I want to know what I’m priced at this weekend.” Until now, that meant you had to open your laptop and do all of that on your desktop. Now you can just open your phone, open your app, and see what you’re priced at for this weekend, for the next weekend, for Christmas.

If something looks off, you can change it right there. You could always do this via the website version on mobile. So you could always open PriceLabs as a website on your mobile. But it’s not the best experience to open a website on a mobile and work through all these graphs. Now that we have built the app, we have actually thought through ensuring that it is mobile responsive. The size of each button, the placement of each button and every screen, is actually very mobile compatible.

Gil: That’s core. For a lot of folks that may not be familiar with mobile development, it’s very attractive to try to cram everything into the mobile app, but the mobile app experience is quite limited in many ways. If you try to do that, you’re going to end up creating a pretty poor experience. So you want to set the expectation on what the mobile app is intended to do. It’s good to know that it was one, quick edits, and two, making sure you’re aware of where your pricing sits.

On the quick edits, there are two specific scenarios I think about often. One is how do I block out my calendar or how do I look at specific dates. And I always forget this one: Thanksgiving lands on a Thursday, and I always have to make sure my pricing and my long stays are accurately priced for Christmas and Thanksgiving every year, because those are our biggest days of the year, especially in the Smokies. That is one where I probably should set it up well in advance so I don’t have to worry about it so much. But that’s one where I think about it last minute, or a friend mentions it to me, like, “Oh, what’s your pricing for this and that?” And I have to look at it.

The other one is if we’re trying to cram in last minute bookings, can I make quick adjustments to provide discounts or change my length of stays, specifically in the 14 day window. We call those last minute. We typically don’t want to get bookings within the 14 days. At that point we’re heavily discounting, so being able to go in there and do it matters. What I don’t want to do is do it at the micro level on the Airbnb side of things, where then all our other channels are out of sync.

Mridul: I agree. Sometimes it’s just that, I remember talking to a host, and she mentioned that out of nowhere she just started getting notifications on her Airbnb that she was getting bookings. And they were four months in advance. She didn’t realize what was happening. Eventually she got to know that it was BTS that were coming to the town, and she was like, “Okay, now it makes sense.” So for instances like these where you weren’t prepared, where you maybe weren’t using a tool to help you with pricing, you can actually take those actions really quickly.

Gil: Looking forward, and you don’t have to share anything that’s in secrecy, is there anything you want to share that you’re working on or continuing to focus your energy toward at a company level? Projects you are excited about that perhaps at the end of this year you’re really starting to ramp up energy toward?

Mridul: Indeed. Directionally what we are trying to do, and that has been our motto since day one, is trying to make revenue management accessible. Now we are leaning toward making it accessible the way you want. Our recent launches have been in that direction. If you want to access revenue management, which means if you want to change your pricing or whatever, you can do it on a mobile app.

Along with the mobile app, we also launched an MCP connector, so you can now use PriceLabs directly with Claude. You don’t even have to log into PriceLabs, you just bring PriceLabs to your Claude. Claude is where a lot of people spend their days now. So you can go to Claude, you can get a summary of what’s happening. You can make changes using Claude, in plain simple language. You don’t even have to go and understand the graphs and charts.

Then we also have our APIs, which we have upgraded. If you use a dashboard or a tool of your own, you can connect PriceLabs to that dashboard and bring both of these together. So you’re getting information, you’re taking actions from your dashboard, but they are actually affecting PriceLabs and your prices and everything connected to that.

This is the direction we are going. Soon we will be launching more such features and products, which will ensure that the effort you have to put in, or the learning curve for anyone to do revenue management, will come down drastically. We are bringing AI everywhere. You can chat with an AI. You can get your things done even without being an expert in all of those things.

Gil: I have a buddy of mine who runs a report every morning. It’s a routine that runs every morning, and it checks his PriceLabs, what it was this year versus the same time last year, what the delta is, any anomalies. It also compares the bookings in the last seven day window.

For him it’s a pulse check to say, “Okay, is everything in a healthy place?” That’s his trigger to say, “Oh, should I be opening PriceLabs right now and looking at things at a granular level?” We’re all juggling a bunch of different things, and as an entrepreneur it’s really about what is the thing you want to tackle next, or what is the thing you need to address. If you don’t have your pulse on things, it’s really hard to know what part of the business you should be spending your energy on this week.

Some people are better where they have cadences, like every Monday they’re doing this, every Tuesday they’re doing this. But inevitably things come along and it changes what our priorities are. Having this routine allows him to see, “Oh no, something’s going off in PriceLabs,” or, “Our booking windows are way shorter than they need to be. This is starting to trend downward.” Claude is really good at identifying some of those things, and now being able to access the data in there, it’s phenomenal.

Mridul: Indeed. You only have 24 hours, and it’s just about how you optimize all of that, how you make the most out of it and prioritize, as you said.

Gil: On the MCP side of it, what’s one prompt that you found, or one thing you can ask your PriceLabs MCP, that you found as a golden nugget? Something that when you found it out, you had to share with others?

Mridul: I think there are a couple. I did post about a few on my personal LinkedIn as well. I was very excited about those. One was related to, a lot of times I see hosts trying to figure out answers to why their price is a certain way, or why their bookings are that way.

That’s the reason we have a few tools within PriceLabs, like the tooltip where you see the breakdown. But one really interesting prompt I found is that you can see why you’re priced lower this weekend. For example, you can say, “Why are my prices down this weekend?” And it will give you a detailed answer about your performance, about the market’s performance, about the thresholds you have added, and it will tell you why.

You can do the same thing with understanding why, and then also what actions you can take. So it can give you suggestions based on those. It can tell you what is happening where, why the market is down, by how much. So I think it’s just very cool.

Gil: I’m guessing what’s happening is that you’re also sending back to Claude what the different signals are. So you’re not just giving the dates and the availability and the rates, but you’re also sending back any spikes in the market, what the demand is in the market, what the booking windows are in the market as well. And then Claude is able to take all that context and put it into a way that you can react to.

Mridul: Exactly, yes. We are sending why, the spikes compared to last year. So why maybe the fifteenth of November 2026 is higher, and we are sending to Claude that it is pacing 10 percent higher than last year. We are sending all of this information so Claude has the context, as you said, to actually answer these questions that you as a host may have.

Gil: That’s very interesting, because I think we’re now at a point where, for a long while, dynamic pricing didn’t exist, or it was only available for hotels and larger operators. But we got to a point where now dynamic pricing is, just like you mentioned, so much more accessible.

Now the differentiator is not necessarily whether or not you have dynamic pricing, because it has in many ways become status quo. There’s now this second level of maturity where it’s not just your dynamic pricing and the tools behind it, but how all your other operations, all your other systems and processes fit together with it. The next tier up in terms of operators doing really well, specifically in revenue management, are the ones that can leverage AI to tap into the revenue management tools and bolster them together.

So you almost have three different tiers, and possibly even more. You have folks that have very flat pricing or are manually adjusting based on their gut. You have folks using dynamic pricing tools within the dashboard. And then you have this third tier where there are property managers using dynamic pricing tools, but they’re also augmenting that with their other AI tools to help them understand what they need to do and how to actually make changes more proactively. It’s almost like entering that gate, and it’s almost a separate tier altogether of maturity for property managers.

Mridul: Indeed, Gil. I totally agree with that. I do want to highlight though that this isn’t something like reaching from tier two to tier three, as you said, that is a necessity. It’s more of how you choose to run your business, how you choose to operate your business.

I know a lot of hosts who just use a dynamic pricing tool, any dynamic pricing tool. They set it right, you provide your thresholds, you provide your preferences. There are a lot of ways to customize tools. You provide, in a way, a download of your business strategy, a download of your preferences, and then just let the automation do the rest.

You don’t come in, you don’t check prices very often. You always have the control and the transparency. You can always go back to it whenever you want. You can always take control from it. But you don’t necessarily have to do anything on a daily basis. I think that works perfectly fine for a lot of people.

The third layer we just talked about is again something which is in case you want to do it, in case that’s how your business operates. Especially for property managers or revenue managers who manage a large number of properties, where it’s literally their job to do it the entire day, it makes more sense for them to leverage AI and make it easier. So it really depends on what size of business you run, what your priorities are, and how much control you want over this part of your business.

Gil: I remember when I first started using PriceLabs in our own portfolio, I had one of my short term rental peers help me set it up because there are a lot of toggles, a lot of switches. This is many years back. I haven’t gone through that process from the infancy level since.

I’m curious on your side, have you changed the onboarding flow based on some of the learnings, and how do you make it easier and more accessible for folks? And perhaps, are you already using or considering using AI to help build that portfolio or that profile? There are a lot of controls you could use, and there’s a basic mode and there’s more of a get into the weeds and fine grain tune things mode. How do you get that person from, “Oh, I’ve been using flat rate pricing for a long time,” or, “I’ve been manually creating things,” to having a pretty solid setup to begin with?

Mridul: That’s a great question, and we definitely think about this a lot. We have people solely working on this part of the product, and it has definitely become easier. Now it is very focused. It’s just a five step journey. You sign up to PriceLabs as someone who is, let’s say, using flat rate pricing. You sign up to PriceLabs. You literally just have to do these five steps, which takes five minutes or ten minutes at max, and you’re all set.

That’s pretty much all you have to do. You come in, and in about three clicks your listings are imported. You just have to set up a base price, which again, we have a tool that helps you with it. So you just have to accept it, in a way. It’s literally just five steps. It takes five to ten minutes, and you’re all set.

What we have also done, and are in the process of doing, is taking away some of those controls and putting them a bit behind the counter, I want to say. In the front, you’ll only see things which matter the most. If you’re someone who actually wants to play with those toggles and customize everything, then you can go to a different section of the product and access all of these things from there. So for someone who doesn’t have a lot of time, who doesn’t want to understand a lot about revenue management because they don’t really have to, they can just do it with these simple toggles, a couple of buttons here and there, and they can get all set up.

About AI, we are in fact working on something which will work as an assistant sort of thing within PriceLabs that you can also use to do all sorts of these things. You can ask it to break down some charts for you. You can ask it about your performance. You can ask it to set up things for you, to make overrides for you. So all of that is again something we are working on.

This is something we genuinely want to continue improving, because we are always at the intersection of wanting to give more controls to the users, more options, but at the same time we should keep it simple as well. That’s the journey we are on. We are figuring it out by keeping some buttons covered in a way where you can go and get them, while keeping the front desk very simple for anyone who wants to get up to speed quickly.

Gil: I think that’s such an important topic in software development. In the past, if you wanted to have advanced capabilities, you would put them behind advanced drop downs or toggles or on different pages. As we’re entering this new wave of AI and chat interfaces, and the ability for computers to infer what we want to do and then synthesize it and produce actions, or even synthesize things that are complex into something understandable, that’s something that wasn’t available for a very long time.

As software changes, we start to go toward, I don’t know if we’ll ever get there or if this is the place we want to be, but this almost headless type of place where you’re not looking at dashboards and you’re using natural language to interface with an application.

At CraftedStays we’ve thought about this a lot, and we’re making a lot of different iterations to figure out the right way to distill down really complex ways to do things. For instance, for us, building a website can be simple, or it can be utterly complex if you really want to get deep into it. On our side, even a few years back, you could create a really beautiful looking website. But as we now go into this new age of AEO and GEO, we’re starting to have a lot more advanced capabilities. We want to bake in more features, but we also want to make sure we don’t stray away from accessibility. We don’t want to make it so complex that people who don’t know about AEO can’t access those features.

What we’re able to do now is leverage chat to really help them synthesize. Right now we launched Aria, for instance, and Aria tells you on your dashboard, “We scanned your portfolio, we scanned the web and figured out what people are searching for. These are the types of content that you’re missing.” Things that typically a human would have to do, an agency you might have hired. It now lowers that barrier to entry, where someone that doesn’t have SEO expertise now knows, “Oh, I should have a landing page on our larger group stays, or our pet friendly properties.”

I think that same thing applies in your space. What I’m hearing from you is that as PriceLabs gets more mature, there are a lot more features, a lot more toggles. Yes, advanced users can go in there and know exactly what to change and how to do things manually, but you also make it accessible so that someone who has a day job doing other things, who is not a revenue manager by trade, can get insights on what their portfolio looks like and react to it, and have almost this conversation where the AI is making all those changes behind the scenes. That’s an amazing place to be in software development right now.

Mridul: Exactly, Gil. And Aria sounds pretty cool. I think that’s very interesting. In marketing this is a big step for people who don’t really want to understand marketing, who are not from this field, but it does affect them at the end. SEO and AEO does affect how you get bookings on your direct booking website. So I think it’s very cool to know about that, that people don’t really have to learn about that anymore. They’ll automatically get tips, I’m assuming, and things like this that they can eventually use to rank better and do all of those things that marketers help you do.

Gil: We’ve already learned. We only launched Aria in earnest to the general public about two months ago. We’ve had beta testers for a long while now, but we’re constantly learning about where people are at. You’re probably seeing the same thing on your side, of what level of sophistication we should bake into Aria. What do people want to do with it? What do they want more of? Because with more sophistication comes more complexity, and then we start to make it less accessible in many ways. So we’re trying to figure out, do we decompose Aria into more bite sized pieces? But it’s a constant pivot. It doesn’t really stop.

Mridul: That’s great. And I agree. Getting customer feedback, actually seeing how people are using it and how they are perceiving it, is definitely the way to go.

Gil: Talk to me a little bit about some of the internal workings. AI has been such a new thing for many folks since 2023. That’s when things started to really pick up. But PriceLabs has been doing AI, actually more specifically ML, machine learning, way before it got really popular, and before chat was the interface people were using to interact with large language models.

Talk to me a bit about some of the data points that PriceLabs uses to help tune and make sure that property managers are maximizing revenue output. What are some of the data points?

Mridul: Definitely. And yes, indeed, I think we’ve been using AI in some form or the other for a long time now. How the data processing works, and I’ll try my best to answer this from a technical perspective, and I hope our actual technical team sees it and agrees with me.

What we do is, it all works on publicly available data of listings on Airbnb, Vrbo, Booking.com, and so on. Our algorithm captures data from these listings around the world for every night. Some data points it picks up are, what’s the ADR? The bookings that are coming in, at what rates are they coming in? What’s the booked price versus the ADR that the host gets?

It sees how many bookings there are. It sees how many active listings there are. For example, in San Francisco in October 2024 there were ten thousand listings on Airbnb, hypothetically. In October 2025 there are eleven thousand listings. Our team has its own definition of what is an active listing, because maybe the listing is active one day for a month versus twenty nine days of a month. So they have their own ways of figuring that out.

Then we look at all of these numbers. We look at how many bookings have come in and how it is pacing. For December 2026, we are already seeing how many bookings have come in the last week for December 2026 versus how many bookings came this week last year for December 2025. So we see how it is pacing ahead or pacing behind.

We get all of this data, and then this data gets processed. That’s where the real quality comes in, because anyone can get this data. It’s publicly available data. Anyone can get this data. The real quality, or the algorithm, comes in refining and making this data usable.

For example, how do you remove owner blocks from it? If someone has just blocked a listing, do you consider that booked or do you consider that blocked? And how do you find that? Because on Airbnb, you will eventually see that as not available. To find that out, and then how do you remove duplicates from it? Maybe one listing is there twice in some way or the other. So this processing of the data you have actually gathered is where the algorithm part, and the AI and the ML, comes in.

All of this processing happens, and then the data is of good quality, I want to say. From there the algorithm picks up different aspects which are important to price your listing. We have different tools. We have dynamic pricing tools, and the job of that is purely to help you find a good price, an optimized price.

Then we have market data tools, let’s say the market dashboard, which is not just for pricing but actually tells you what’s happening in your market. And then we have revenue estimation tools, which tell you how much potential you can earn from a property or a location.

So based on the tool, based on the need that the user has, different parts of those tools and those data points that have been collected get shown to you. The algorithm picks up these things and uses them to, if I talk about dynamic pricing, recommend a price for your property.

And it picks up signals. It doesn’t even have to know the cause. I’ve seen this happen a few times where we don’t even know what’s happening on that date. We just know the demand is high. For example, the BTS tour that I just told you about, our algorithm may not even know that BTS is coming there. The algorithm would say, “Okay, prices for this weekend are up by 60 percent for some reason.” That is an indicator that, wherever it applies, based on every user’s own preferences, the prices for this weekend should go up. Eventually it recommends a price which is aligned with that demand and data that they are seeing.

Maybe later on someone will realize, okay, BTS is coming, but that doesn’t mean you have to wait to know what is there to actually get the right price. So that’s in a nutshell how that works.

Then the algorithm finds a price, and it looks at your own performance as well. I think one of the important parts of dynamic pricing is that it doesn’t work just based on market data. If you were to just blindly follow market data, I don’t think that’s the way it works. It also understands your occupancy and your performance in general.

So maybe for the month of December you are booked, your occupancy is 80 percent, whereas last year in December your occupancy was 90 percent. Clearly you would understand this as a down period where you are making less than you were, or you’re less booked than you were last December. But when you look at market data, the market is actually down by 20 percent. So you being down 10 percent is actually better. You’re still outperforming the market.

That’s where your factor comes in as well. Let’s say for next weekend the market is quite down, but you’re already doing good enough. You’re already good as per your past data and your market. So your prices don’t necessarily have to go down. Even if the market is down, if you are doing good enough, you can still stay at the same price and just hope to get booked. Otherwise, you’re still doing good enough. So this is where that algorithm part really comes in.

The algorithm finds a price, and it sees how you are doing. And the third and most important part of it is your preferences. Everyone wants to run their business their own way. Let’s say I prefer to have 100 percent occupancy. I prefer to be booked as many days as I can for the month, even if I’m making thin margins. Even if I’m just covering my cost, even if I’m just making a couple of dollars on it, I’m okay with it, because otherwise I’m not making anything at all.

Whereas someone who is also living in that same property, who is maybe renting a part of the property, may have a preference where if it gets booked it’s a bit of a discomfort for them to have someone come over and stay at their property. Maybe their preference is, if my property gets booked, I want a good rate, otherwise I’m okay with keeping it empty as well.

So all of these preferences, how aggressive you want to be with your pricing, how safe you want to play with your pricing. These three factors, which are market data, your performance, and your preferences, all three combined generate a price which the tool recommends to you at the end.

Gil: So if I captured that correctly, it’s a combination of what the market aggregate data is, how your peers in your space are doing. You’re looking at the historicals of how this year compares to last year. Are we trending down? Are we trending up? Are there macro shifts in the overall rental market?

Your own performance, specifically the occupancy, how you’re trending, what your pacing looks like, what your booking windows look like. How do you stack up against the rest of the market? And then the last one is your preferences, your risk tolerance, the sensitivity to the data being presented.

Those are the things that ultimately help decide what that nightly rate is, and also how length of stay is calculated. Did I capture that right? Historical data, aggregate market data, your own performance, and then also your preferences.

Mridul: Exactly. You got it exactly right, Gil.

Gil: It’s always been this black box. I’m in these Facebook groups and I see people vibe coding all these different things. One of the things I saw, one of the property managers vibe coded their own dynamic pricing tool. Like you mentioned, some of this data is publicly available, so they’re scraping at a micro level in their specific market, and then they’re basically trying to build their own dynamic pricing tool.

My first immediate reaction is that it makes me cringe a little bit, because it can be simplified to these points, but really understanding this is at a much higher level. One thing we didn’t talk about is that because PriceLabs has such market momentum, there are so many property managers using it, and you can see at a much more granular level what price is being set, how other property managers are performing, that is not publicly available data. I’m assuming that data is also being used to tweak the end result as well, and you don’t have that own data if you were to build your own solution.

There are a couple of things that folks should be arming themselves with and learning how to use Claude and other AI tools. But there are things that folks should not try to replicate. Dynamic pricing is one of them. Your property management system is another one. And to some degree, even your own website too. I’ve had a lot of folks that are really proud of the vibe coded website they share. They’ll post it on Facebook. And when I review those websites, I look under the hood at what’s actually happening, how the website’s being built, and I’m like, “You’re missing a lot of the fundamentals that go behind it.” So it’s very easily trivialized, what you can do versus what you should do, and what you should really rely on the platforms to take care of.

Mridul: Just one thing that I would like to clarify, or rather change, is that in terms of data, you can actually build it. All the data we use is publicly available. So even if you were to vibe code, you can do it. We do not use one customer’s data to price another customer.

We don’t use any property manager who’s using PriceLabs, it doesn’t matter for our data. We actually just end up using publicly available data. So even if only ten property managers were using PriceLabs today versus, I think, 80,000 now, the recommendations that PriceLabs would give would be exactly the same. Or the model, rather, would be exactly the same. Because what we are getting is public data, which is already available.

So if anyone wants to vibe code, yes, they can do it, and they can get all the data that we have as well. It’s just about the processing that I said. It’s all about how do you find these nuances, which our team has learned over the last 12 years now.

It’s actually about finding those nuances, because a lot goes into it, which to be frank, even I don’t understand completely. It’s the engineers that we have. We have some of the best engineers in the world in terms of data. The algorithm that we have has actually won awards.

So it’s really about treating that data. It’s not just about gathering the data. Anyone can have the data. It’s really about treating that data right. Finding duplicates, finding blocked nights, finding the actual ADR. Because as platforms are changing the way that the nightly rate is shown to the guest, let’s say all inclusive rates and things like these, how do you work around all of that to actually find what the price is?

Let’s say now you get the nightly rate, the booked rate. But then you need to separate out the commission that was paid to Airbnb, and then how do you bring it back to your platform so that it gets applied? Otherwise, let’s say yesterday a property got booked for $300. Today I can’t recommend $300, because the $300 also had Airbnb’s commission in it. How do you account for all of these things and then create all of this and give that?

Another part that I really like about the algorithm is how it is hyperlocal. It really captures and considers the properties and competitors and market data that matter to you. If there’s a lake or a river, and properties are on both sides of it, if you were to just map out a radius, you would most likely end up including the properties on the other side as well. But they are not really your competitors, because people who are staying here are not even considering that, and vice versa.

So how do you figure out this hyperlocal market, which is finding the data, finding the competitors, finding the market that is really applicable to you, and not just a blanket market of your entire city? If Taylor Swift is performing in San Francisco, the rates of properties which are very close to the stadium versus properties which are 10 miles away will be different. How do you find that geography? How do you find that area which affects your property and eventually your rates? That is equally important, because data is there. Data is there. How do you work around the algorithm to actually find data that is actually relevant to you?

Gil: I like the lake example. We have a client that is on Lake Anna, actually several clients on Lake Anna. Lake Anna has two sides. You have the warm side and the cold side. The warm side is the private side of the lake where it’s not open to the public, and you have the cold side.

The prices between the two sides of the lake are dramatically different. It’s a different experience. And it’s not something that even shows up in the amenities. There’s no warm side of the lake dropdown amenities box that you can check in your PMS. So it is a lot of geography based there. It could be private. There’s maybe some amenity flag for that.

You’re right. If you were to try to build your own dynamic pricing tool and you’re using some sort of radius to determine who your competitors are, you could be comparing properties on the warm side of the lake that have private access versus the public side, which is dramatically different in terms of pricing. You can be calibrating your pricing completely off.

Mridul: Totally agree. It’s all about finding the relevancy and what matters to your property.

Gil: Mridul, this has been phenomenal. I really love all the nuggets of information, having you explain how things have evolved at PriceLabs, and getting into some of the projects you really enjoy, like launching the mobile app, and touching upon some of the MCPs and what that enables for folks.

We usually end the show with three questions. The first question is, what’s a book that has inspired you, that you have an affinity toward?

Mridul: There are so many, but if I have to pick one, I would say Shoe Dog by Phil Knight. It’s not related to the industry that I work in, it’s not related to the kind of work that I do. But that’s the book that I read when I was quite young, and that’s something that has stuck with me. I think that’s the book that had the most impact on me in some sort of way.

Gil: Is that the book that’s based on the story of Nike and how it became what it is today?

Mridul: Indeed, yes. It’s a memoir by Phil Knight, the founder of Nike, and it’s a beautiful story. He really shows what it takes to build a company. Starting from the struggles and how he eventually got to build what he did, and the hardships, the down moments throughout his journey, personal losses and whatnot. I think it’s really a story which really inspires me.

Gil: Nice. Second question. What’s one piece of mindset advice that you would give to someone that’s starting something completely new?

Mridul: I would say do it for the process, which is, enjoy the process. As long as you’re enjoying the process, I think the outcome doesn’t matter. So if you’re doing something new, don’t do it just because there will be a golden day, or just because you will eventually get something out of it. Yes, that’s the goal, and you should try to work toward it, but you really should enjoy what you’re doing every day to achieve that. If you want to build a business, if you want to do anything of that sort, really enjoy what it is taking, and every piece, every bit of time that you spend working on it.

Gil: I love that. Last question. For any of our listeners that are looking to either get started in direct bookings or amplify their direct bookings, what’s a tactical tip they can put into action this afternoon to start that?

Mridul: I would say don’t think of it as OTAs versus direct bookings, or OTAs or direct bookings. I think it’s always a combination of both. One analogy that I generally like is to use OTAs as billboards. The billboard effect, for people who may not know, is essentially that you never get customers directly from billboards. People see a billboard of your company, your product, your services, and eventually they learn about it and buy from some channel or the other.

Use OTAs to be that billboard. If you want to focus on direct bookings, then still be attractive on OTAs. Still be competitively priced there. Still have good pictures there, a good listing description there, all of those things, so that people find you, they find you attractive there, and then they come to your website to maybe get a better deal.

Now, you give a better deal. It can be in terms of pricing, it can be in terms of an add on that you do. One thing I’ve seen people do is, if you know your guest is celebrating something special, maybe put a bottle of wine there when they come in, or maybe send them flowers or something like that. Just go one step above, which doesn’t really have to cost too much. It’s just a gesture.

And then also build a bridge between these two. For example, have the name of your brand in your listing description, or even the title if you can, so that it becomes easy for someone to find you. Imagine, show your OTA listing to a stranger and then ask them to find your website. How much time does it take for them to find your direct booking website? So if you’re attractive there, if you’re a good billboard, if you’re giving a good deal on your website, and if you’re building a bridge which people can walk on from OTA to your website, I think you’ll be doing a pretty good job.

Gil: I love that. That’s something very tactical that I think every property manager getting started in direct booking should be doing today. Putting your brand name either in the descriptions, I’ve seen some people put it in the titles themselves, some people will even embed it into the pictures. There are some rumors about whether that is a good thing or a bad thing to do. Definitely don’t put it in your messages. Avoid putting that in your messages. But definitely build a brand out there so that folks can find you by whatever means.

I had a guest that wanted to book with us and they had a very particular budget, and our properties are not the cheapest properties out there. But they had a budget and they really wanted our stay. We told them, “Sorry, at Explore Stay Today, we can’t discount our taxes, we can’t discount the Airbnb fees.” And they were able to pick up on the notes that, “Oh, I can book with you outside of this.” They sent me an email and said, “I’ll book with you directly.”

Because the fees were so large, they were able to get it under the budget they wanted and still get the experience. And honestly, I much prefer to have folks book with us directly, because I find that they are just nicer guests when they’re not on the Airbnb platform. I never feel like they’re trying to swindle me for a refund. We have pretty good checks in place to make sure that we don’t bring in the wrong type of guests. But that was an example of using the billboard effect quite strategically in your messaging.

I totally agree with you, the OTAs are a great place to do marketing in the right way. And it’s a great place to get started.

Mridul: Indeed. I agree. That’s a beautiful story.

Gil: Mridul, it was a huge pleasure having you on the show and having you share more tactically what’s happening within PriceLabs and some of the things that you think about. So I appreciate it. If folks want to learn more about PriceLabs or want to follow you, where can they reach you?

Mridul: Folks can definitely visit our website, which is hello.pricelabs.co. They can learn more about PriceLabs. They can sign up for a free trial. We don’t charge anything for you to test things out. There’s a 30 day free trial. If anyone wants to reach out to me, they can reach out to me on my LinkedIn.

Gil: Awesome. I’ll be sure to include that in the show notes, and again, thanks for joining us today.

Mridul: Thanks, Gil. Thanks for having me.